Approximate Triple Modular Redundancy: A Survey

In recent years, approximate computing (AC) has attracted attention owing to its tradeoff between the exactness of computations and performance gains. AC has also been probed for the technique of Triple modular redundancy (TMR). TMR is a well-known fault masking methodology, with associated overheads, widely used in systems of different nature and at different levels. E.g.: layout-level, gate-level, HW-module level, software. At hardware level, through exploitation of AC the 200% area overhead problem due to triplication of the original modules in TMR can be reduced. By approximating the modules of TMR while ensuring that at least two of the approximate modules do not differ from the original module for every input vector, the facilitation of fault masking can lead to overhead reduction. Hence, approximate TMR (ATMR) aims to achieve cost-effective reliability. Nevertheless, due to the extensive search space, computational complexity, and principal fault masking function of ATMR, designing an ATMR is a challenging task. An ATMR technique must be scalable so that it can be easily adopted by circuits having large number of inputs and the extraction of ATMR modules remains computationally inexpensive. Compared with TMR, due to the inclusion of approximations, ATMR is more vulnerable to errors, and hence, the design technique must ensure awareness of input-criticality. To the best of the authors’ knowledge, none of the existing survey articles on AC has reported on ATMR. Therefore, in this work, ATMR design techniques are thoroughly surveyed and qualitatively compared. Moreover, design considerations and challenges for designing ATMR are discussed.

[1]  Sparsh Mittal,et al.  A Survey of Techniques for Approximate Computing , 2016, ACM Comput. Surv..

[2]  Antonio Martínez-Álvarez,et al.  Improving approximate-TMR using multi-objective optimization genetic algorithm , 2018, 2018 IEEE 19th Latin-American Test Symposium (LATS).

[3]  Mayler G. A. Martins,et al.  Building ATMR circuits using approximate library and heuristic approaches , 2019 .

[4]  Farshad Baharvand,et al.  LEXACT: Low Energy N-Modular Redundancy Using Approximate Computing for Real-Time Multicore Processors , 2020, IEEE Transactions on Emerging Topics in Computing.

[5]  Antonio Martínez-Álvarez,et al.  On the influence of compiler optimizations in the fault tolerance of embedded systems , 2016, 2016 IEEE 22nd International Symposium on On-Line Testing and Robust System Design (IOLTS).

[6]  Server Kasap,et al.  Survey of Soft Error Mitigation Techniques Applied to LEON3 Soft Processors on SRAM-Based FPGAs , 2020, IEEE Access.

[7]  Mayler G. A. Martins,et al.  Exploring the use of approximate TMR to mask transient faults in logic with low area overhead , 2015, Microelectron. Reliab..

[8]  Mario García-Valderas,et al.  Logic masking for SET Mitigation Using Approximate Logic Circuits , 2012, 2012 IEEE 18th International On-Line Testing Symposium (IOLTS).

[9]  Jeong-A Lee,et al.  Probing Approximate TMR in Error Resilient Applications for Better Design Tradeoffs , 2016, 2016 Euromicro Conference on Digital System Design (DSD).

[10]  A. Bosio,et al.  Approximate TMR based on successive approximation and loop perforation in microprocessors , 2019, Microelectronics Reliability.

[11]  Fernanda Gusmão de Lima Kastensmidt,et al.  Reducing TMR overhead by combining approximate circuit, transistor topology and input permutation approaches , 2013, 2013 26th Symposium on Integrated Circuits and Systems Design (SBCCI).

[12]  Luis Entrena,et al.  Approximate TMR for selective error mitigation in FPGAs based on testability analysis , 2018, 2018 NASA/ESA Conference on Adaptive Hardware and Systems (AHS).

[13]  Antonio Martínez-Álvarez,et al.  Design of approximate-TMR using approximate library and heuristic approaches , 2018, Microelectron. Reliab..

[14]  Mayler G. A. Martins,et al.  Methodology for achieving best trade-off of area and fault masking coverage in ATMR , 2014, 2014 15th Latin American Test Workshop - LATW.

[15]  Jeong-A Lee,et al.  A Fault Tolerant Voter for Approximate Triple Modular Redundancy , 2019, Electronics.

[16]  Alberto Bosio,et al.  Survey on Approximate Computing and Its Intrinsic Fault Tolerance , 2020, Electronics.

[17]  Bernard Girau,et al.  Fault and Error Tolerance in Neural Networks: A Review , 2017, IEEE Access.

[18]  Karthikeyan Sankaralingam,et al.  Dark Silicon and the End of Multicore Scaling , 2012, IEEE Micro.

[19]  David W. Coit,et al.  Multi-objective optimization using genetic algorithms: A tutorial , 2006, Reliab. Eng. Syst. Saf..

[20]  Alberto Bosio,et al.  Exploiting Approximate Computing for Low-Cost Fault Tolerant Architectures , 2019, 2019 32nd Symposium on Integrated Circuits and Systems Design (SBCCI).

[21]  Roberto R. Osorio,et al.  Truncated SIMD Multiplier Architecture for Approximate Computing in Low-Power Programmable Processors , 2019, IEEE Access.

[22]  Khaled Salah Mohamed.  Approximate Computing: Towards Ultra-Low-Power Systems Design , 2020 .

[23]  Jörg Henkel,et al.  Design Automation of Approximate Circuits With Runtime Reconfigurable Accuracy , 2020, IEEE Access.

[24]  L. Entrena,et al.  Partial TMR in FPGAs Using Approximate Logic Circuits , 2015, IEEE Transactions on Nuclear Science.

[25]  Mayler G. A. Martins,et al.  Using only redundant modules with approximate logic to reduce drastically area overhead in TMR , 2015, 2015 16th Latin-American Test Symposium (LATS).

[26]  Akash Kumar,et al.  A flexible inexact TMR technique for SRAM-based FPGAs , 2016, 2016 Design, Automation & Test in Europe Conference & Exhibition (DATE).

[27]  Luis Entrena,et al.  Error Mitigation Using Approximate Logic Circuits: A Comparison of Probabilistic and Evolutionary Approaches , 2016, IEEE Transactions on Reliability.

[28]  Qiang Xu,et al.  Guest Editors' Introduction: Approximate Computing , 2016, IEEE Des. Test.

[29]  John P. Hayes,et al.  Selective Hardening: Toward Cost-Effective Error Tolerance , 2011, IEEE Design & Test of Computers.

[30]  Jeong-A Lee,et al.  Input vulnerability-aware approximate triple modular redundancy: higher fault coverage, improved search space, and reduced area overhead , 2018 .